To date, owing to their wide availability,1.5T systems are used for the majority of fMRI studies although high-field (3T and 4T) systems have also been used. Conceptual considerations of BOLD mechanism predict a fundamental dependence on magnetic field strength and led to the suggestion that the sensitivity, contrast, and spatial specificity of the BOLD response to neural activity increase with the field strength. These predictions, however, have only been examined so far in a relatively few studies all based on animal models. While animal studies provide us with data that can elucidate the biophysics of fMRI in certain models, they are not necessarily fully applicable to the human brain BOLD fMRI. First, most animal studies are conducted under anesthesia, where the physiological parameters are different from that of awake human subjects. Second, the signal-to-noise ratio as well as spatial and temporal resolutions achievable with animals are much more than that achievable in humans, making the translation of animal results to humans inappropriate. Third, paradigms that can be applied to animal models are mostly limited to sensory stimulation. Fourth, the vascular architecture in animals differs from that in humans with respect to all vessels except capillaries. For these reasons, it is important to investigate fMRI in humans at ultra high magnetic fields to elucidate the field dependence of various attributes of fMRI and ascertain the advantages of high magnetic field. With the availability of a 7 Tesla whole-body imager,it is now possible to investigate these issues directly in the human brain for the first time. The overall goal of the present application is to investigate the characteristics of fMRI at ultra high magnetic fields and utilize these potential advantages for high resolution fMRI that can probe neuronal function at the millimeter to sub-millimeter spatial scale. As ultrahigh field human systems are still in their experimental stage, technical development is needed to make use of its capabilities. To realize the anticipated increase in sensitivity and spatial specificity, it isnecessary to develop methodology for high-resolution fMRI at ultrahigh fields. In addition, as physiological noise may scale with the signal in the data, become dominant, and mitigate the advantages of high field, it is important to understand it and to develop improved methods to reduce it. Thus the first aim of this project will focus on these two technical aspects. Our second aim will then investigate the field dependence of sensitivity and specificity and how these issues affect spatial and temporal resolutions. For these studies we propose to focus on high resolution fMRI of brain regions rather than the whole brain at this stage since the latter goal presents an additional set of challenges that are beyond the scope of this application. With these considerations in mind, our specific aims are 1) technical development and understanding of signal fluctuations in fMRI at high fields and 2) characterization of the BOLD response at ultrahigh magnetic field with high spatial and temporal resolution.